In the most simplistic sense, the abuse of alcohol is related to its availability, i.e., if there is no alcohol available, no consumption occurs and no abuse is possible. However, the number of factors which affect or influence consumption and abuse are multiple. Many studies of alcohol abuse show positive correlation between availability and abuse. While it is unlikely that correlational findings are causal, that is, availability directly causing abuse, there do exist positive relationships. These relationships are most likely complex networks of interrelated variables which affect one another over time. There have been few research efforts to examine the complex relationships between availability and abuse as a total or "whole." This project seeks to solve these limitations by undertaking a three-year study of the relationships between availability and abuse through the use of a computer-based model which will integrate in a wholistic manner the most significant variables. The focus of this project is to explore the feasibility of and to test the capability of a computer-based model to incorporate the best available knowledge and experience about alcohol availability and abuse. The project will carry out sensitivity analysis with the model to locate those variables which are most sensitive to change over time (i.e., would be most affected by changes in other variables), and to carry out simulation experiments to plot the projected effects of changes in the significant and most powerful variables in the model and to prepare reports on findings.